Published online 13 July 2005
Original article
Lung cancer mortality in European women: recent trends
and perspectives
C. Bosetti
1*, F. Levi
2, F. Lucchini
2, E. Negri
1& C. La Vecchia
1,31Istituto di Ricerche Farmacologiche ‘Mario Negri’, Milan, Italy;2Registre Vaudois des Tumeurs, Institut Universitaire de Medecine Sociale et Preventive,
Lausanne, Switzerland;3Istituto di Statistica Medica e Biometria, Universita degli Studi di Milano, Milan, Italy
Received 7 March 2005; revised 4 May 2005; accepted 27 May 2005
Background:Lung cancer mortality in men has been declining since the late 1980s in most European countries. In women, although rates are still appreciably lower than those for men, steady upward trends have been observed in most countries. To quantify the current and future lung cancer epidemic in European women, trends in lung cancer mortality in women over the last four decades were analyzed, with specific focus on the young.
Patients and methods:Age-standardized (world standard) lung cancer mortality rates per 100 000 women—at all ages, and truncated 35–64 and 20–44 years—were derived from the WHO for the European Union (EU) as a whole and for 33 separate European countries. Joinpoint regression analysis was used to identify points where a significant change in trends occurred.
Results:In the EU overall, female lung cancer mortality rates rose by 23.8% between 1980–1981 and 1990–1991 (from 7.8 to 9.6/100 000), and by 16.1% thereafter, to reach the value of 11.2/100 000 in 2000–2001. Increases were smaller in the last decade in several countries. Only in England and Wales, Latvia, Lithuania, Russia and Ukraine did female lung cancer mortality show a decrease over the last decade. In several European countries, a decline in lung cancer mortality in young women (20–44 years) was observed over the last decade.
Conclusions:Although female lung cancer mortality is still increasing in most European countries, the more favorable trends in young women over recent calendar years suggest that if effective interventions to control tobacco smoking in women are implemented, the lung cancer epidemic in European women will not reach the levels observed in the USA.
Key words:Europe, lung cancer, mortality, trends, women
Introduction
Lung cancer mortality in men has been declining since the late 1980s in most European countries. The overall fall was over 13% (from 55.4 to 46.7/100 000 men, world standard) between 1988 and 2000 in the 25 countries of the European Union (EU) [1]. Downward trends were observed also in the Russian Feder-ation, and central and eastern European countries, characterized by exceedingly high rates in the early 1990s [2–4]. The only exceptions were Portugal and Romania, where male lung cancer mortality was still increasing.
In women, lung cancer mortality rates in most Europe are still appreciably lower than those for men, but the pattern of trends is largely different, since steady upward trends have been observed in most countries, reaching extremely high levels (20 to 25/
100 000 at all ages) in Denmark, Iceland, Ireland and the UK [3–7]. In most other European countries, female lung can-cer mortality rates are still below 10/100 000, substantially lower than in North America [8, 9]. A clearer understanding of the ongoing lung cancer epidemic among European women requires detailed analysis of trends in separate age groups.
We present therefore a comprehensive analysis of trends in lung cancer mortality in European women over the last four decades, with specific focus on the young, who are of specific interest to shed light on the most likely future trends [10, 11].
Patients and methods
Official death certification numbers for lung cancer for 33 European coun-tries (including the Russian Federation, but excluding a few small councoun-tries such as Andorra and Liechtenstein) for the period 1965–2001 were derived from the WHO database as available on electronic support [12]. As well as the UK as a whole, data are also presented separately for England and Wales, Scotland and Northern Ireland. Data for the Russian Federation and other Republics included in the former Soviet Union have been available in *Correspondence to: Dr C. Bosetti, Laboratorio di Epidemiologia,
Istituto di Ricerche Farmacologiche ‘Mario Negri’, Via Eritrea 62, 20157 Milan, Italy. Tel: +39-02-39014526; Fax: +39-02-33200231; E-mail: bosetti@marionegri.it
the WHO database from the early 1980s onwards. For Albania, data are available only since 1989, for Belgium up to 1997, and for Denmark, France and The Netherlands up to 2000.
The EU was defined as the 25 member states as in May 2004 (i.e. Austria, Belgium, the Czech Republic, Cyprus, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, The Netherlands, Poland, Portugal, Slovakia, Slovenia, Spain, Sweden, UK). Data for Cyprus were not available. During the calendar period considered (1965–2001) three different revisions of the International Classification of Diseases (ICD) were used [13–15]. Classification of cancer deaths was recoded, for all calendar periods and countries, according to the Ninth Revision of the ICD [14].
Estimates of the resident population, generally based on official censuses, were obtained from the same WHO database [12]. From the matrices of certified deaths and resident populations, age-specific rates for each 5-year age group and calendar year were computed. Age-standardized rates per 100 000 women—at all ages, and truncated 35–64 years and 20–44 years—were computed using the direct method, and based on the world standard population [16].
Joinpoint regression analysis was performed using the joinpoint software from the Surveillance Research Program of the US National Cancer In-stitute [17]. This analysis allows us to identify points where a significant change in the linear slope of the trend occurred [18]. In joinpoint analysis, the best fitting points (the ‘joinpoints’) are chosen where the rate changes significantly. The analysis starts with the minimum number of joinpoints (e.g. zero joinpoints, which is a straight line), and tests whether one or more joinpoints (up to three) are significant and must be added to the model. In the final model, each joinpoint (if any) indicates a significant change in the slope. The estimated annual percent change (EAPC) is then computed for each of those trends by fitting a regression line to the natural logarithm of the rates using calendar year as a regressor variable [i.e. given y = a+ bx, where y = ln(rate) and x = calendar year, the EAPC is estimated as: 100 * (ebÿ 1)].
Results
Table 1 gives the age-adjusted mortality rates from lung cancer in women (at all ages, and truncated 35–64 and 20–44 years) for 33 European countries in 1980–1981, 1990–1991 and 2000– 2001, and the corresponding percentage changes between the three periods. In the EU as a whole, female lung cancer mortal-ity rose by 23.8% between 1980–1981 and 1990–1991 (from 7.8 to 9.6/100 000). A further increase, though smaller (16.1%), was observed afterward, to reach the rate of 11.2/ 100 000 in 2000–01. Steady upward rates between 1980–1981 and 1990–1991 were observed in most European countries, with the highest rises found in some countries of northern, central and eastern Europe, including Denmark, Germany, Hungary, Nor-way and Poland. Smaller increases were found in several coun-tries, with the exception of France and Switzerland, which showed larger increases between 1980–1981 and 1990–1991. Only in Latvia, Lithuania, England and Wales, the Russian Fed-eration, and Ukraine did female lung cancer mortality show a decrease between 1990 and 2001. In England and Wales, however, lung cancer rates in 2001 were still20/100 000.
Lung cancer mortality in EU middle-aged women increased by 17.2% between 1980–1981 and 1990–1991 (from 7.8 to 9.6/100 000), and by a further 21.1% thereafter, to reach a value of 17.9/100 000 in 2000–01. In most European countries, the
changes in truncated rates between 1980–1981 and the 1990s were similar to those for the overall ones, the only exceptions being Ireland, the Russian Federation, Ukraine, and England and Wales, which already showed declines in rates. Between 1990–1991 and 2000–2001 rates in middle-aged women showed a decline in a few European countries, including Britain, Denmark, Ireland, the Russian Federation and Ukraine. In France and Spain, however, steady upward trends were observed in this period.
We also considered trends in young women (20–44 years) from various European countries in 1980–1981, 1990–1991 and 2000–2001 (Table 1). In the EU, lung cancer mortality rates per 100 000 in young women were 1.1 in 1980–1981, 1.5 in 1990–1991 and 1.7 in 2000–2001, corresponding to percentage increases of 35.5 and 11.4% in the two periods, respectively. In various European countries, however, a decline in lung cancer mortality in young women was observed over the last decade.
Figure 1 shows trends in female lung cancer mortality at all ages and truncated at 35–64 years for selected European coun-tries between 1965 and 2001. A steady upward trend in overall rates was observed in several European countries over the four decades considered. A levelling of lung cancer mortality was observed in Denmark, Iceland (since the late 1990s), Ireland and the UK (since the late 1980s). In these countries, however, rates had steeply increased in the previous decades, when they reached the highest values on a European scale. Trends in middle-aged women follow similar, though more pronounced, patterns than those for all women. Thus, the rises were stronger than those in overall rates in most countries with increasing mortality from lung cancer. In countries where overall lung cancer mortality tended to level off in more recent years (i.e. Denmark, Iceland, Ireland and the UK), truncated rates showed declining trends.
Lung cancer mortality trends in young women (20–44 years) over the period 1965–2001 in selected European countries are shown in Figure 2. Although less clear than overall and 35–64 years trends due to smaller numbers of deaths, the trends over recent calendar periods appeared to be more favorable than those in all-age and middle-aged women in most European countries. Thus, in several European countries (including Austria, Hungary, Italy, The Netherlands, Poland, Sweden and Switzerland), rates tended to decline in young women in the last few years. Steady long-term declines in rates were ob-served in Ireland and the UK since the late 1960s. Lung cancer mortality rates in young women were still consistently increas-ing in France and Spain up to more recent calendar years.
The main findings from jointpoint regression analysis for fe-male lung cancer mortality (at all ages, and truncated at 35–64 and 20–44 years) over the period 1965–2001 in the EU overall are given in Table 2 and Figure 3. At all ages, female lung cancer rates rose by 1.5% per year between 1965 and 1971, by 2.6% between 1972 and 1987, and by 1.4% thereafter. In middle-aged women, the rise was 2.1% per year until 1983, 1.4% between 1984 and 1997, and 2.9% thereafter. In young women, after a decrease between 1965 and 1969 (ÿ0.8% per year), rates rose by 2.8% per year between 1970 and 1997, and subsequently declined by ÿ3.6% over the last 4 years.
the European Union, in the years 1980–81, 1990–91 and 2000–01
Country All ages Age 35–64 years Age 20–44 years
Death rate/100 000 Percent change Death rate/100 000 Percent change Death rate/100 000 Percent change
1980–81 1990–91 2000–01 1990–91/ 1980–81 2000–01/ 1990–91 1980–81 1990–91 2000–01 1990–91/ 1980–81 2000–01/ 1990–91 1980–81 1990–91 2000–01 1990–91/ 1980–81 2000–01/ 1990–91 Albania – 5.07a 6.72 – 32.50 – 7.39a 8.15 – 10.3 – 1.31a 1.39 – 6.1 Austria 7.31 9.19 11.11 25.7 20.90 10.19 14.21 19.22 39.5 35.3 0.75 2.09 1.71 178.7 ÿ18.2 Belgium 6.48 8.48 10.02b 30.9 18.20 10.47 15.09 16.64b 44.1 10.3 1.14 1.89 1.87b 65.8 ÿ1.1 Bulgaria 5.69 6.36 6.38 11.8 0.30 8.88 10.15 11.3 14.3 11.3 1.32 1.74 1.61 31.8 ÿ7.5 Croatia – 8.19 10.76 – 31.40 – 13.92 16.97 – 21.9 – 2.02 1.91 – ÿ5.4 Czech Republic – 9.91 12.75 – 28.70 – 16.88 20.58 – 21.9 – 1.57 1.15 – ÿ26.8 Denmark 15.35 24.88 27.44c 62.1 10.3 29.31 44.14 38.96c 50.6 ÿ11.7 2.27 2.13 2.48c ÿ6.3 16.4 Estonia 7.53d 7.57 7.49 0.5 ÿ1.1 9.23d 11.73 10.64 27.1 ÿ9.3 0.72d 1.11 1.05 54.2 ÿ5.4 Finland 5.99 6.85 7.86 14.4 14.7 8.49 8.70 10.52 2.5 20.9 0.66 0.75 0.71 13.6 ÿ5.3 France 3.84 5.19 7.31c 35.2 40.9 6.07 8.41 13.43c 38.6 59.7 0.61 1.08 2.12c 77.1 96.3 Germany 5.65 8.10 10.75 43.4 32.7 8.62 12.94 18.21 50.1 40.7 0.82 1.41 1.64 72.0 16.3 Greece 6.03 7.22 7.56 19.7 4.7 9.33 11.51 10.99 23.4 ÿ4.5 1.54 1.43 1.16 ÿ7.1 ÿ18.9 Hungary 10.07 15.73 21.25 56.2 35.1 16.23 28.15 42.06 73.4 49.4 1.90 4.89 4.96 157.4 1.4 Iceland 23.46 23.44 26.68 ÿ0.1 13.8 32.93 39.21 39.32 19.1 0.3 Ireland 15.70 17.74 17.80 13.0 0.3 26.30 25.60 19.33 ÿ2.7 ÿ24.5 2.41 0.94 1.34 ÿ61.0 42.6 Italy 6.05 7.32 8.38 21.0 14.5 9.50 10.62 12.17 11.8 14.6 0.92 1.11 1.24 20.7 11.7 Latvia 5.7d 6.73 6.06 18.1 ÿ10.0 9.85d 9.63 8.65 ÿ2.2 ÿ10.2 0.72d 0.69 0.73 ÿ4.2 5.8 Lithuania 4.99d 6.09 5.34 22.0 ÿ12.3 8.99d 9.33 6.94 3.8 ÿ25.6 0.33d 0.96 0.75 190.9 ÿ21.9 Luxembourg 6.69 9.13 10.94 36.5 19.8 10.42 15.51 19.99 48.9 28.9 Malta 5.54 2.82 7.13 ÿ49.1 152.8 5.88 3.38 9.85 ÿ42.5 191.4 The Netherlands 6.29 10.53 16.87c 67.4 60.2 11.11 20.16 30.72c 81.5 52.4 1.28 2.04 2.63c 59.4 28.9 Norway 5.81 10.81 16.32 86.1 51.0 10.35 19.03 25.53 83.7 34.2 0.86 1.34 1.67 55.8 24.6 Poland 7.05 10.05 12.69 42.6 26.3 12.46 17.6 22.75 41.3 29.3 1.20 1.89 1.81 57.5 ÿ4.2 Portugal 3.73 4.53 4.79 21.5 5.70 5.91 6.72 7.43 13.7 10.6 1.13 1.3 1.18 15.0 ÿ9.2 Romania 6.22 6.49 7.89 4.34 21.3 11.38 11.79 13.53 3.6 14.8 1.76 1.64 1.77 ÿ6.8 7.9 Russian Federation 6.53 7.30 5.95 11.8 ÿ18.5 11.08 10.99 8.86 ÿ0.8 ÿ19.4 1.31 1.2 1.10 ÿ8.4 ÿ8.3 Slovakia – 7.33e 7.61 – 3.8 – 11.99e 12.02 – 0.3 – 1.42e 1.43 – 0.7 Slovenia – 8.31 10.83 – 30.3 – 14.36 18.42 28.3 28.3 – 0.73 2.20 – 201.4 Spain 3.83 3.61 4.71 ÿ5.7 30.5 5.53 5.3 8.77 ÿ4.2 65.5 0.91 1.15 1.67 26.4 45.2 Sweden 7.67 10.35 13.84 34.9 33.7 13.05 17.91 22.30 37.2 24.5 1.32 1.67 1.06 26.5 ÿ36.5 1599
Discussion
The present comprehensive analysis of female lung cancer mor-tality in Europe indicates and quantifies a steady increase in rates in women from most European countries over the last few decades. In the EU overall, the rates increased from 5.53/ 100 000 in 1965 to 11.16 in 2001 at all ages, and from 9.15 to 17.88 in middle-aged women. However, as in the USA [19], a gradual slowing in the rate of increase has been observed over recent years. According to the jointpoint regression analysis, female lung cancer mortality in the EU increased by 2.6% per year between the early 1970s and the late 1980s, and by 1.4% thereafter.
Rises in female lung cancer rates were consistently observed in several European countries, although the degree of the increase varied between countries. Thus, particularly sharp increases were observed in some northern, central and eastern countries such as Denmark, Germany, Hungary and Poland. In England and Wales, Ireland and Iceland [5], where rises were earlier and mortality rates were higher, a levelling off in overall rates—and a decline in middle-aged and young women—was observed in more recent years.
The different patterns of female lung cancer mortality reflect the different prevalence of tobacco smoking in women from various European countries, which in turn can help in interpret-ing recent and future trends in lung cancer mortality. The lag in the temporal trend of lung cancer mortality rates in women compared with men also reflects historical differences in ciga-rette smoking between men and women. In some northern coun-tries such as Belgium, Denmark, Sweden and The Netherlands, where the estimated smoking prevalence in women has fallen recently, a slowing of lung cancer mortality rates can be expected in the future, as already observed in the UK [20–22]. The female lung cancer epidemic seems to be still in its early phases in countries like Austria, Spain and France, where smok-ing prevalence in women has been rapidly increassmok-ing in the 1990s [21, 22]. Among countries of central and eastern Europe, Hungary is the only country where a steady increase in smoking prevalence has been observed since the 1960s [23]. Conversely, the prevalence of tobacco smoking among Russian, as well as Ukrainian, women has remained relatively low in the last dec-ades (i.e.10–15%) [24]. The declining trends observed in the Russian Federation should be taken with caution since they are also influenced by effects in earlier cohorts, owing to the limited availability of cigarettes in generations who were teenagers in the post-war period [25].
A major finding of the present analysis is the more favorable lung cancer mortality trends in young women, particularly in countries where a peak has already been reached, suggesting that overall trends are likely to be more favorable in the future. Trends for young adults are in fact an early indicator of the recent and potential future impact of changes in the prevalence of risk factors—notably tobacco smoking—on cancer rates [11, 26]. Also in the USA, the effect of decreasing the prevalence of smoking was seen first in young adults [27–29]. We chose to use the age-standardized rates at age 20–44 years as a measure of lung cancer trends in the young, as suggested by Doll in the
T able 1. (Continued ) Count ry All ages Age 35–6 4 year s Age 20–4 4 years Death rate /100 000 Perce nt change Death rate/1 00 000 Percent change Death rate/1 00 000 Per cent change 1980 –81 1990 –91 2000–01 1990 –91/ 1980 –81 2000 –01/ 1990 –91 1980–81 1990 –91 2000 –01 1990–91 / 1980–81 2000–01 / 1990–91 1980–81 1990–91 2000 –01 1990 –91/ 1980 –81 2000 –01/ 1990 –91 Switzerland 5.43 7.50 10.70 38.1 42.7 9.15 12.8 4 18.3 3 40.3 42.8 1.23 1.62 1.25 31.7 ÿ 22.8 Ukraine 6.55 d 7.35 5.51 12.2 ÿ 25.0 11.7 d 11.2 3 8.97 ÿ 4.0 ÿ 20.1 1.61 d 1.42 1.31 ÿ 11.8 ÿ 7.8 UK 17.8 2 21.0 6 19.75 18.2 ÿ 6.2 30.24 29.3 9 24.7 4 ÿ 2.8 ÿ 15.8 1.78 1.56 1.19 ÿ 12.4 ÿ 23.7 UK, Engl and/W ales 17.4 0 20.3 7 18.83 17.1 ÿ 7.6 29.27 28.2 7 23.6 7 ÿ 3.4 ÿ 16.3 1.71 1.48 1.13 ÿ 13.5 ÿ 23.7 UK, N orthern Ireland 14.1 6 17.1 4 19.66 21.1 14.7 25.70 24.8 6 24.0 8 ÿ 3.3 ÿ 3.1 1.04 1.42 0.87 36.5 ÿ 38.7 UK, Scot land 22.8 0 28.8 4 28.92 26.5 0.3 40.86 41.2 35.6 0.8 ÿ 13.6 2.62 2.40 1.84 ÿ 8.4 ÿ 23.3 Europ ean Unio n 7.76 9.61 11.16 23.8 16.1 12.61 14.7 8 17.9 17.2 21.1 1.10 1.49 1.66 35.5 11.4 aData refer to the year s 1988 –198 9. bDa ta refer to the ye ars 1996 –1997. cData refer to the year 2000 . dDa ta refer to the ye ar 1981 . e Data refer to the year s 1992 –199 3. 1600
Figure 1. Trends in age-standardized (world population) death certification rates per 100 000 for lung cancer in women (all ages, and age 35 to 64 years) in selected European countries, 1965–2001. All ages, solid lines; truncated at 35–64 years, dotted line.
Figure 2. Trends in age-standardized (world population) death certification rates per 100 000 for lung cancer in women aged 20–44 years in selected European countries, 1965–2001.
1990s [11], and adopted for instance by Polednak [29] to ana-lyze lung cancer incidence trends in black and white young adults in the USA. These rates are heavily influenced by the last quinquennia of age, since over 90% of all lung cancer deaths at age 20–44 years occur at 35–44 years, and over 70% at 40–44 years (median age at death 42 years). Other measures of lung cancer rates in the young have been suggested, such as using the age group 30–39 years [28]. These are, however, on average at a younger age (median age at death 38 years), and are based on smaller absolute numbers. Therefore, this would be a major problem for smaller countries.
Thus, since EU female lung cancer rates at age 20–44 years have leveled off in the late 1990s at values50% lower than those of their male counterparts, it is likely that overall lung cancer rates in EU women will continue to increase for some years, to then stabilize at a value 15/100 000 between 2015 and 2020. Any more precise estimate is, however, hampered by major uncertainties in the prevalence of smoking in women over the next few years, and mostly by the role of stopping smoking over the next decades [30]. In the presence of effective inter-vention to reduce smoking among European women, the peak rate may be lower. Assuming a constant 1.4% rise per year between 2000 and 2015, the rate would approach 14/100 000. With a 1% rise, it would remain13/100 000.
In the interpretation of the present results it is important to consider problems related to random variation, which are clearly greater in relation to smaller populations. Secondly, and more complex, there are problems of death certification reliability and validity in various countries [31, 32]. In general, for lung cancer, death certification is sufficiently reliable to permit meaningful inference on trends for most European countries. Furthermore, trends in the young are less likely to be affected by certification problems. Some under-recording of cancer deaths was reported for the Russian Federation in the late 1980s and 1990s, due to a fall in precision of coding of causes of death. This was, how-ever, mainly restricted to the elderly living in rural areas [33], and should therefore not have materially influenced rates at younger ages. Moreover, no major changes in lung cancer treat-ments and survival have occurred in the last decades that could have materially influenced mortality trends [34].
Overall, age-standardized female lung cancer mortality across Europe is still much lower than in the USA, where lung cancer has become the leading cause of cancer death among women, with a rate of 24/100 000 in the year 2000 [8,9]. Only a few European countries have female lung cancer rates comparable to those of the USA, but there a peak seems to have been reached. In most other European countries, the lower extent of more re-cent increases compared with those of the past, and the more favorable trends in young women, suggest that female lung cancer mortality rates will probably not reach the high levels observed in the USA [5, 8]. Effective interventions to control and reduce tobacco smoking in women should be implemented to avoid a major lung cancer epidemic in European women in the near future.
Table 2. Age-standardised (world population) lung cancer mortality rates per 100 000 women and joinpoint analysis (1965–2001), at all ages, truncated at 35–64 years and at 20–44 years in the European Union
Age-standardized mortality rates Jointpoint analysis
1965 2001 Trend 1 Trend 2 Trend 3
Years EAPC Years EAPC Years EAPC
All ages 5.53 11.16 1965–1971 1.51 1972–1987 2.59a 1988–2001 1.39a
35–64 years 9.15 17.88 1965–1983 2.08a 1984–1997 1.39a 1998–2001 2.93a
20–44 years 1.19 1.59 1965–1969 ÿ0.84a
1970–1997 2.78a 1998–2001 ÿ3.60a aSignificantly different from 0 (P <0.05).
EAPC, estimated annual percent of change.
0 5 10 15 20 1965 1970 1975 1980 1985 1990 1995 2000 Calendar year
Death rate per 100,000
Death rate per 100,000
a
0.5 1 1.5 2 1965 1970 1975 1980 1985 1990 1995 2000 Calendar year 0b
Figure 3. Jointpoint analysis for female lung cancer mortality in the European Union, 1965–2001. (A) All ages, squares; truncated at 35–64 years, circles. (B) Age 20–44 years.
Acknowledgements
The authors thank Mrs M. P. Bonifacino for editorial assistance. This work was conducted with the contribution of the Italian and Swiss Leagues against Cancer, the Swiss Foundation for Research against Cancer, and the Italian Association for Cancer Research.
References
1. Levi F, Lucchini F, Negri E, La Vecchia C. Trends in mortality from major cancers in the European Union, including acceding countries, in 2004. Cancer 2004; 101: 2843–2850.
2. Levi F, Lucchini F, Negri E, La Vecchia C. The end of the tobacco-related lung cancer epidemic in Europe. J Natl Cancer Inst 2003; 95: 631–632.
3. Levi F, Lucchini F, Negri E et al. Cancer mortality in Europe, 1995– 1999, and an overview of trends since 1960 (published erratum appears in Int J Cancer 2004; 111: 981). Int J Cancer 2004; 110: 155–169. 4. Levi F, Lucchini F, Negri E et al. Trends in cancer mortality in the
European Union and accession countries, 1980–2000. Ann Oncol 2004; 15: 1425–1431.
5. Levi F, La Vecchia C, Lucchini F, Negri E. Lung cancer in Icelandic women. Eur J Cancer Prev 1999; 8: 369.
6. Borras JM, Fernandez E, Gonzalez JR et al. Lung cancer mortality in European regions (1955–1997). Ann Oncol 2003; 14: 159–161. 7. Bray F, Tyczynski JE, Parkin DM. Going up or coming down?
The changing phases of the lung cancer epidemic from 1967 to 1999 in the 15 European Union countries. Eur J Cancer 2004; 40: 96–125. 8. Jemal A, Chu KC, Tarone RE. Recent trends in lung cancer mortality
in the United States. J Natl Cancer Inst 2001; 93: 277–283. 9. Bosetti C, Malvezzi M, Chatenoud L et al. Trends in cancer mortality
in the Americas, 1970–2000. Ann Oncol 2005; 16: 489–511. 10. Doll R, Peto R. The causes of cancer: quantitative estimates of
avoid-able risk of cancer in the United States today. J Natl Cancer Inst 1981; 66: 1191–1308.
11. Doll R. Progress against cancer: an epidemiologic assessment. The 1991 John C. Cassel Memorial Lecture. Am J Epidemiol 1991; 134: 675–688.
12. World Health Organization Statistical Information System. WHO mor-tality database. http://www3.who.int/whosis/menu.cfm. Geneva: World Health Organization 2004.
13. World Health Organization. International Classification of Disease: 8th Revision. Geneva: World Health Organization 1967.
14. World Health Organization. International Classification of Disease: 9th Revision. Geneva: World Health Organization 1977.
15. World Health Organization. International Statistical Classification of Disease and related Health Problems: 10th Revision. Geneva: World Health Organizatio 1992.
16. Doll R, Smith PG. Comparison between registries: age-standardized rates. In: Waterhouse JAH, Muir CS, Shanmugaratnam K et al. (eds):
Cancer Incidence in Five Continents, Vol. IV. IARC Sci Publ 1982; No. 42. Lyon: International Agency for Research on Cancer 1982; 671–675. 17. National Cancer Institute. Joinpoint Regression Program, version 2.7.
September 2003. http://srab.cancer.gov/joinpoint
18. Kim H-J, Fay MP, Feuer EJ, Midthune DN. Permutation tests for join-point regression with application to cancer rates (published erratum appears in Stat Med 2001; 20: 655). Stat Med 2000; 19: 335–331. 19. Howe HL, Wingo PA, Thun MJ et al. Annual report to the nation on the
status of cancer (1973 through 1998), featuring cancers with recent increasing trends. J Natl Cancer Inst 2001; 93: 824–842.
20. Graham H. Smoking prevalence among women in the European community 1950–1990. Soc Sci Med 1996; 43: 243–254.
21. European Network for Smoking Prevention. Some like it light. Women: smoking in the European Union. ENSP Report 1998.
22. Molarius A, Parsons RW, Dobson AJ et al. and WHO MONICA Project. Trends in cigarette smoking in 36 populations from the early 1980s to the mid-1990s: findings from the WHO MONICA Project. Am J Public Health 2001; 91: 206–212.
23. Tyczynski JE, Bray F, Aareleid T et al. Lung cancer mortality pat-terns in selected Central, Eastern and Southern European countries. Int J Cancer 2004; 109: 598–610.
24. Gilmore A, Pomerleau J, McKee M et al. Prevalence of smoking in 8 countries of the former Soviet Union: results from the living condi-tions, lifestyles and health study. Am J Public Health 2004; 94: 2177– 2187.
25. Shkolnikov V, McKee M, Leon D, Chenet L. Why is the death rate from lung cancer falling in the Russian Federation? Eur J Epidemiol 1999; 15: 203–206.
26. Muir CS, Fraumeni JF Jr, Doll R. The interpretation of time trends. Cancer Surv 1994; 19–20: 5–21.
27. Wingo PA, Ries LAG, Giovino GA et al. Annual report to the nation on the status of cancer, 1973–1996, with a special section on lung cancer and tobacco smoking. J Natl Cancer Inst 1999; 91: 675–690. 28. Jemal A, Cokkinides VE, Shafey O, Thun MJ. Lung cancer trends in
young adults: an early indicator of progress in tobacco control (United States). Cancer Causes Control 2003; 14: 579–585.
29. Polednak AP. Lung cancer incidence trends in black and white young adults by gender (United States). Cancer Causes Control 2004; 15: 665–670.
30. Peto R, Darby S, Deo H et al. Smoking, smoking cessation, and lung cancer in the UK since 1950: combination of national statistics with two case–control studies. BMJ 2000; 321: 323–329.
31. Percy C, Stanek E 3rd, Gloeckler L. Accuracy of cancer death cer-tificates and its effects on cancer mortality statistics. Am J Public Health 1981; 71: 242–250.
32. Boyle P. Relative value of incidence and mortality data in cancer research. Recent Results Cancer Res 1989; 114: 41–63.
33. Shkolnikov V, McKee M, Vallin J et al. Cancer mortality in Russia and Ukraine: validity, competing risks and cohort effects. Int J Epidemiol 1999; 28: 19–29.
34. Spira A, Ettinger DS. Multidisciplinary management of lung cancer. N Engl J Med 2004; 350: 379–392.
